Generated test data and published example data for Neuro-MINE protocol

Generated test data These data can be used to test basic functionality of Neuro-MINE. Their structure is described in Costabile, J. D., Balakrishnan, K. A., Schwinn, S., & Haesemeyer, M. (2023). Model discovery to link neural activity to behavioral tasks. Elife, 12, e83289. fit_test_predictors.csv: Predictor file for testing Neuro-MINE. fit_test_responses.csv: Corresponding response file MINE_fit_test_responses_run_config.json: Run configuration to produce expected outputs MINE_fit_test_responses_Insights.csv: Expected outputs from the run. Mouse example data These data are from one session of one mouse from Musall, S., Kaufman, M. T., Juavinett, A. L., Gluf, S., & Churchland, A. K. (2019). Single-trial neural dynamics are dominated by richly varied movements. Nature neuroscience, 22(10), 1677-1686. These examples are used for the figures in the Neuro-MINE protocol manuscript. mSM30_episodic_predictors.zip: Predictor data from one session, broken down into individual trials or episodes. This produces slightly better fits than treating the data as continuous. For a simpler, non-episodic test, consider using the data from version 1 of this repository which combined the trials in a single file. mSM30_responses.zip: The corresponding wide-field imaging components, broken down into individual trials or episodes. MINE_mSM30_epidosid_run_config.json: Run configuration to produce expected outputs MINE_mSM30_episodic_Insights.csv: Expected outputs from the run. Please note that some units are close to the fit threshold and stochastic variation in the fit can therefore lead to slight changes from run to run.